DESIGN OPTIMIZATION WITH ADVANCED GENETIC SEARCH STRATEGIES

The present paper describes the capabilities of a modern design optimi zation tool based on the method of genetic search. This stochastic sea rch technique offers a significantly increased probability of locating the global optimum in a design space with multiple relative optima. T he program includes an advanced search technique referred to as direct ed crossover wherein bit positions on the design strings that offer a higher gain during crossover are assigned higher probabilities of sele ction as crossover sites. This optimization code also includes a multi stage genetic search plan that is useful in problems where the design space is large. Multistage search involves successive refinement in th e precision with which design variables are represented in the genetic search process. Also included in this program is a newly developed cl uster identification technique that automatically determines the cente r location and the radius of a hypersphere representing a relative-opt imum containing region. Cluster information serves to define accurate parameters required for other advanced techniques such as sharing func tion implementation and mating restrictions.